Excel Spreadsheet Handler

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script recalc.py executes the system binary soffice (LibreOffice) in headless mode using subprocess.run to perform formula recalculations on Excel files.- [COMMAND_EXECUTION]: The skill automatically generates and writes a StarBasic macro file (Module1.xba) into the user's LibreOffice configuration directory (~/.config/libreoffice/ or ~/Library/Application Support/LibreOffice/) to enable recalculation functionality.- [PROMPT_INJECTION]: The skill processes untrusted Excel files via pandas and openpyxl, creating a surface for indirect prompt injection where malicious content within spreadsheet cells could influence the agent's behavior during analysis.
  • Ingestion points: pd.read_excel() and load_workbook() calls in SKILL.md used for document analysis.
  • Boundary markers: No specific delimiters or instructions are provided to ensure the agent ignores potentially malicious text embedded within cells.
  • Capability inventory: The skill possesses file system read/write access and the ability to execute system commands through the recalc.py script.
  • Sanitization: No documented sanitization or validation of cell content is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 12:30 PM